OBJECTIVE: Intraarticular screw perforation is a common complication of open reduction and internal fixation of proximal humerus fractures. The purpose of this study was to (1) determine the sensitivity and specificity of the fluoroscopic images used to evaluate whether a screw tip has perforated into the glenohumeral joint, and (2) determine the specific fluoroscopic views that best evaluate screw position in the humeral head. METHODS: Twenty-two proximal humeri in 11 lightly embalmed cadavers were instrumented. The articular surface was divided into equal-sized rows (superior, central, inferior) and columns (anterior, middle, posterior). The screws in 10 humeri were inserted and so their tips were located 2 mm beneath the articular surface. Twelve humeri had screws placed such that their tips protruded 2 mm past the articular surface into the glenohumeral joint. Twenty-seven different C-arm views were obtained of each specimen/screw configuration. RESULTS: There were zero false-positives (100% specificity). The average sensitivity was 55% and varied greatly depending on the image view and the screw exit location (range, 0%-100%). The sensitivity for the inferior row of screws was the lowest (39.1%) and was particularly low for the posterior-inferior screw exit location (20.7%). CONCLUSIONS: Screws that are completely located within the bone of the proximal humerus will never appear on C-arm images as intraarticular. However, screws that are intraarticular may appear to be completely located within the bone of the proximal humerus on some C-arm images. A sensitivity of 100% for detecting intraarticular screws for 8 of the 9 screws' exit locations and 90% for the posterior-inferior screw can be achieved by imaging the proximal humerus in 25-degree internal rotation, neutral, and 25-degree external rotation with the C-arm in neutral cant at rainbow 25-degree roll over, neutral rainbow, and rainbow 25-degree roll back for a total of 9 images.
OBJECTIVE: Intraarticular screw perforation is a common complication of open reduction and internal fixation of proximal humerus fractures. The purpose of this study was to (1) determine the sensitivity and specificity of the fluoroscopic images used to evaluate whether a screw tip has perforated into the glenohumeral joint, and (2) determine the specific fluoroscopic views that best evaluate screw position in the humeral head. METHODS: Twenty-two proximal humeri in 11 lightly embalmed cadavers were instrumented. The articular surface was divided into equal-sized rows (superior, central, inferior) and columns (anterior, middle, posterior). The screws in 10 humeri were inserted and so their tips were located 2 mm beneath the articular surface. Twelve humeri had screws placed such that their tips protruded 2 mm past the articular surface into the glenohumeral joint. Twenty-seven different C-arm views were obtained of each specimen/screw configuration. RESULTS: There were zero false-positives (100% specificity). The average sensitivity was 55% and varied greatly depending on the image view and the screw exit location (range, 0%-100%). The sensitivity for the inferior row of screws was the lowest (39.1%) and was particularly low for the posterior-inferior screw exit location (20.7%). CONCLUSIONS: Screws that are completely located within the bone of the proximal humerus will never appear on C-arm images as intraarticular. However, screws that are intraarticular may appear to be completely located within the bone of the proximal humerus on some C-arm images. A sensitivity of 100% for detecting intraarticular screws for 8 of the 9 screws' exit locations and 90% for the posterior-inferior screw can be achieved by imaging the proximal humerus in 25-degree internal rotation, neutral, and 25-degree external rotation with the C-arm in neutral cant at rainbow 25-degree roll over, neutral rainbow, and rainbow 25-degree roll back for a total of 9 images.
Authors: George D Chloros; Apostolos D Prodromidis; Jo Wilson; Peter V Giannoudis Journal: Eur J Trauma Emerg Surg Date: 2021-09-06 Impact factor: 2.374
Authors: Jan Theopold; Kevin Weihs; Christine Feja; Bastian Marquaß; Christoph Josten; Pierre Hepp Journal: BMC Med Imaging Date: 2017-08-01 Impact factor: 1.930